I WOULD LIKE to respond to a question posed by the Managing Director in his opening remarks as to why the Fund has found it so difficult to make pronouncements on the exchange rates of major currencies. Some ten years ago the Fund did not hesitate to make such a pronouncement. For example, in August 1971 the staff felt sure enough of its ground to propose a new set of exchange rates. In the event, the rates proposed proved wrong: the changes were too small and did not produce enough adjustment. There was, however, a rather comforting explanation: it was not that the Fund’s model was wrong but that the size of the 1971 disequilibrium had been seriously underestimated. In the mid-1970s, the Fund staff started to calculate “underlying balances” for the major countries—i.e., balances that would materialize over the medium term with present exchange rates and at reasonably high levels of activity in all countries. The exercise proved generally right in predicting the fall in the deutsche mark/dollar rate from 1976 to September 1978 and it also correctly suggested that the further sharp fall of the dollar in October 1978 went too far. But the model was never good at predicting what would happen to the yen, and it broke down for the dollar in 1980. Since then, interest rates have been so dominant and so volatile over so long a time, with such wide-ranging effects, and with so much bandwagon riding in both the capital and the foreign exchange markets, that the kind of medium-term analysis in which the Fund had engaged has become largely irrelevant to what happens in exchange markets for the short term and also to what can be made to happen through, for example, intervention or any policy of bands or target zones.
In commenting on the conference, I must confess to having a slight bias in favor of ideas in the areas of policy rather than of analysis. However, I appreciate the middle course that Mr. Branson has steered between very simple and very complicated models. By a modest addition to oversimple models, he has introduced quite realistic aspects of countries’ economic structure, such as the commodity content of imports and exports, nominal and real wage rigidity, and the presence or absence of effective financial markets. I would like to emphasize two aspects of the Branson approach. First, what may appear as structural characteristics of an economy to the economic analyst need not necessarily be considered in this way for policy purposes. It was true, for example, that full indexation of money wages and the consequential extreme stickiness of real wages were notable characteristics of the economies of many European countries. If one accepts this structural element as immutable, exchange rate policies—and also many other policies of adjustment—become totally ineffective, and any form of adjustment that presupposes the reduction of real wages is by definition impossible. The task of policy in these circumstances then becomes not one of accommodating exchange rate policy to the structure but one of changing the structure so as to make exchange rate policy effective. This indeed has been the main issue of adjustment in recent years in a number of European countries, such as Belgium, France, Ireland, Italy, and the Netherlands, and the key to successful policies in these countries.
Second, Mr. Branson has introduced financial markets as a structural difference between industrial and developing countries. However, I believe that he assumed too readily that the financial markets in the former countries are thick enough for the monetary authorities to trust private speculation to be stabilizing, and, hence, to permit reliance on floating rates. My own impression is that confidence in stabilizing capital movements has gone out of fashion in the last few years, for a number of good reasons, of which Mr. Dornbusch has briefly mentioned the prevalence of irrelevant information, the possibility of changes in the regime that encouraged speculators to hold out for a break in the system, and “bandwagon effects”—all three of which have devastating consequences for the attainment of equilibrium. Quite apart from the trust that can be put in the resilience of financial markets, there was more to structure on this point than whether countries export manufactures or primary products. There is also the need to separate the industrial countries according to their size. There has been remarkably little reference during the day to the concept of the optimum currency area, even though the comparison of the costs and benefits of floating versus fixed rates has a lot to do with the size and openness of a country. Moreover, as we become increasingly less sanguine about the stabilizing nature of speculation, the balance of benefits versus costs as a function of economic size will inevitably shift further in the direction of fixed against floating rates.
One point that has been stressed throughout the day is that, whatever their exchange rate regime, small industrial countries are inevitably exposed to the main economic events occurring in their larger neighbor countries. Should these countries try to exploit whatever freedom they have to control their own business cycle or their own interest rates by following a separate exchange rate policy? Or should they come to terms with the degree of their economic and financial dependence and opt for a fixed rate against the currency of their largest neighbor? A number of medium-sized countries seem to have come to quite different answers to this question; for example, Canada has opted for a floating rate for its currency vis-á-vis the U.S. dollar, while the Netherlands has chosen a fixed rate on the deutsche mark.
I find an absence of dogmatism in today’s presentations on fixed versus floating rates and a wide area of parallelism of findings; this is shown by the extent to which all four authors tended to dissent from Harry Johnson’s view that flexible rates are essential to the preservation of national autonomy and “independence consistent with efficient organization and development of the world economy.” Specifically, Mr. Arriazu has stressed that variations in output and employment have tended to be much larger under floating than under fixed rates, and Mr. Swoboda has observed that floating has not permitted countries to avoid riding, or rather being ridden by, the world business cycle. However, I think it is important not to forget the lesson from the 1930s, when those countries that lowered their exchange rates (such as Japan, Sweden, and the United Kingdom) did notably better in real terms than some of the other main industrial countries. This may have been because nominal depreciations in the 1930s invariably produced real effective depreciations, whereas nowadays they often do not. Or it might be another case of confusion of terminology between the general impact of floating and the specific impact of depreciation.
I hope that economists will have a clear idea of what is involved in making a comparative judgment between fixed and flexible exchange rate systems, at least in particular country situations. I must caution against a lowering of standards in making such judgments. Viability, for instance, is being treated as one criterion. The par value system is not viable in the face of severe policy conflicts, as the events of 1971 and 1973 proved. In a narrow sense, a floating rate system is viable under any circumstances. But the experience of Mexico in the last few weeks has illustrated how shallow this viability can be. I support Mr. Swoboda’s conclusion that any judgment about the merits of floating rates must be based on their success in terms of other variables, such as inflation, unemployment, cyclical stability, and adjustment. A certain lowering of standards is also visible in the European Monetary System. What else is the rejoicing about the system’s ability to survive successive realignments in disregard of the fact that it is the absence of convergent policies—that is, the failure of the European Monetary System in its basic objective—that made these realignments necessary?
I support Mr. Branson’s view that we should not look to “better policies” and “coordination” to solve all problems of international policy. Countries have some national instruments, such as the exchange rate, that give them some freedom from the effects of other countries’ policies. However, there are limits to the use of any instrument. Living as we do in a period of severe shocks, world inflation, and distortion in the income distribution, is there any point in looking for an exchange regime that will let us get through this period without very unpleasant consequences? Countries cannot escape the consequences of their own policies through their selection of an exchange rate regime.